Introduction. The concept of race and prostate cancer. JW Moul 1 * Keywords: prostate; cancer; screening; black race; PSA

Transcription

1 (2000) 3, 248±255 ß 2000 Macmillan Publishers Ltd All rights reserved 1365±7852/00 $ Targeted screening for prostate cancer in African-American men{ 1 * 1 Urology Service, Department of Surgery, Walter Reed Army Medical Center, Washington, DC and Center for Prostate Disease Research, Uniformed Services University of the Health Sciences, Bethesda, MD, USA African-American men and black men throughout the world have a higher rate of prostate cancer than other ethnic groups. They also are most likely to present at a younger age with more advanced disease and have historically had a poorer prognosis. Whether this observed difference is due to behavior, lack of access, environmental factors or genetics is hotly debated. Whatever the cause or causes, there is growing concensus that targeting screening to this high-risk group is justi ed. Focused education about risk and screening in black men can be effective and demonstration screening programs in African-American community settings have been successful. There is much debate about the proper normal values of PSA to be used in screening high risk black men. Some have argued for a very low normal range such as 2.0 ng/ml to improve sensitivity for detection of curable disease. Others, recognizing the balance between sensitivity (cancer detection) and speci city (avoiding unnecessary prostate biopsies), have proposed age-adjusted PSA ranges. Until randomized or prospective screening trials are completed, it seems reasonable to encourage black men to start screening at age 40 using a PSA threshold of 2.0 to 2.5 ng/ml to prompt further evaluation. Prostate Cancer and Prostatic Diseases (2000) 3, 248±255. Keywords: prostate; cancer; screening; black race; PSA Introduction The age-adjusted incidence of prostate cancer in African American (black) males, is 50% higher than in Caucasian (white) men, and African-American (black) men have the highest incidence of prostate cancer in the world. 1 Furthermore, the mortality rate is 55.1/ for black men compared to 24.7/ for white men, a two-fold difference, indicating that prostate cancer is a public health problem in this population. 2 The etiology for these racial differences in prostate cancer clinical manifestations is unknown; hormonal, nutritional, genetic, behavioral and socioeconomic status (SES) factors have all been implicated. 3 Despite the etiological debate, there is emerging knowledge that focused education can *Correspondence:, Center for Prostate Disease Research, 1530 E. Jefferson Street, Rockville, MD 20852, USA. jmoul@cpdr.org y The opinion and assertions contained herein are the private views of the author and are not be construed as re ecting the views of the US Army of the Department of Defense. Received 8 May 2000; accepted 24 August 2000 improve early detection efforts in this high-risk group. Furthermore, there is now a large body of literature about prostate-speci c antigen (PSA) and ne-tuning its use in African-American men. This current report will update these important areas that are critical to urologic practice. The concept of race and prostate cancer The concept of race is considered a social de nition taking into account historical, social and economic events not a biological classi cation. 5 Some have argued for abandoning race in variable analysis because it has no biologic meaning and is medicalization of racism. 6,7 Recently, Brawley and Freeman contend that equal treatment yields equal outcome among patients with the same stage of disease regardless of race. 8 They further argue that race is not a biologic category and that the concept of biologically more aggressive tumors in blacks be abandoned. However, the fact remains that African-American or black men have a greater risk of being diagnosed with prostate cancer and their disease is usually more advanced and potentially more aggressive. The simple

2 goal of research in this area is to improve outcomes for these men. Realizing that race may be an indicator of economic status, cultural preferences or misunderstandings, and genetic susceptibility for cancer behavior and development, the pragmatic mission is to understand these contributing factors to prostate cancer in this highrisk group and eliminate the observed disparity. Knowledge and attitudes Conventional wisdom has been that African-American men are less knowledgeable and more reluctant to undergo prostate cancer screening than other groups; however, a number of recent studies have shed more light on this issue. Robinson et al conducted focus groups on prostate cancer screening in middle and lower socioeconomic groups of African-Americans. 9,10 Among middle socioeconomic participants there was a willingness to undergo screening and individuals possessed reasonable knowledge; however, among the low socioeconomic groups, there were many misconceptions and myths about the etiology and mortality of cancer. Fewer of the low socioeconomic men knew that prostate cancer was a greater problem in blacks than whites and many were unaware of rectal examination or blood tests. There was concern in the low socioeconomic men that blacks are `used as guinea pigs' in reference to the Tuskegee syphilis trials. There was also fear of hearing bad news and misconceptions of surgery causing cancer to spread and rectal examination having homosexual implications. 9 The authors felt that culturally relevant education and establishing a patient ± provider relationship of trust and respect were needed. Abbott et al also conducted a study of 944 Caucasian and African-American men regarding prostate knowledge. 11 At baseline, the black men were signi cantly less likely to identify early symptoms of prostate cancer than white men, but after an educational session there were no racial differences except that blacks still felt that `pain' was the rst symptom of prostate cancer. Despite this improvement in knowledge with education, Bennett et al have documented the impact of literacy. 12 In 212 lowincome men, blacks were almost twice as likely to present with metastatic prostate cancer (49.5% vs 35.9%; P < 0.05) and have literacy less than sixth grade level (52.3% vs 8.7%; P < 0.001) compared to whites and literacy but not race was associated with stage. Guidry et al have studied educational materials for prostate cancer and found that only 40% are culturally sensitive, taking into account low literacy and other readability factors. 13 Myers et al 14 and Weinrich and associates 15 have documented that African-American peer-education and tailored behavioral intervention can improve prostate education and early detection efforts. In this light, the recently reported Detroit Education and Early Detection (DEED) study of Powell and associates must be detailed. 16 The DEED program conducted prostate cancer education and screening in the Detroit area, focusing on African-American churches, and screened 1105 black men between 1993 and Eighty- ve of the 1105 men (8%) had a PSA value of greater than 4.0 ng/ml and 36 cancers (3.2% overall, 42% of men with elevated PSA) were detected. Although a success, the program was only able to reach the middle income (median $30 943) and educated (average of `some college'). The lost to followup rate (11.7%) and biopsy refusal rate (4.7%) in DEED was felt to be not signi cantly higher than screening programs in white populations. In the screened DEED men who underwent radical prostatectomy, the rate of organ-con ned disease was 65%, which is higher than historic controls of black surgical patients. 16,17 Considering the relatively high detection rate in this young (mean age 54.9 y) group of black middle socioeconomic men, further efforts to reach the low socioeconomic and even younger men seem most warranted. 18,19 Screening for prostate cancer in African-American men Early detection for prostate cancer has been practiced for many years in the form of digital rectal examination (DRE). However, over the last decade with the advent of PSA, the topic of prostate cancer screening has become a hotly contested issue. 20,21 A full discussion of populating-based screening is beyond the scope of this current report; however, the key question is whether screening black men should be considered `population-based' or treated more like `case- nding' in a high-risk group similar to those men with a family history of the disease. The latest guidelines from the American College of Physicians 22 and the American Cancer Society 23 dictate against routine screening but recommend apprising men of the pros and cons of testing and letting the wellinformed individual decide. While theoretically sound, in practical terms many physicians do not take the time or have the time for such counseling and the screening may or may not be done based on the bias of the practice. Furthermore, in light of new `stage migration' data in the PSA era 24,25 and even one randomized trial reporting a bene t to population-based screening, 26 the pros and con discussion is controversial. Speci cally, should clinicians give the `bene t-of-doubt' that screening may improve long-term outcome or emphasize the uncertainty. The dilemma is more dif cult for African-American men: we want to lessen the disparity with early detection but we currently lack de nite date on screening ef cacy in this high-risk group. In the DEED program, Powell et al found that screened African-American men had a higher likelihood of organ-con ned, nonadvanced prostate cancer. 16 Furthermore, encouraging data from the Radiation Therapy Oncology Group (RTOG), 27 US Military 28 and Veterans Administration, 29 suggest that if black men are afforded the same access, are diagnosed with early stage disease, and receive similar care, the outcomes disparity may be minimized or eliminated. Screening and early detection issues in African-American men remain. The proper age to initiate testing is debated. While Powell et al feel that screening in black men should start at an earlier age, 30 the proper age between 35 ± 45 and even up to age 50 is not known and deserves study. The issue of PSA in African-American men is widely debated and is discussed below. 249

3 250 PSA in black men In 1992, Vjayakumar and associates were the rst to report that black American men with newly diagnosed prostate cancer referred for radiotherapy had higher PSA levels than their white counterparts. 31 A number of other preliminary early reports also suggested that blacks had higher PSA. 32,33 These early reports lacked proper multivariate adjustment for stage, grade are and socioeconomic status. In 1995, our group reported on 541 consecutive men with newly diagnosed prostate cancer and showed that, even with adjustment for tumor grade, age and clinical stage, black men had higher PSA. 34 In this same study, we took a subsequent consecutive cohort of 91 black and white men who underwent a radical prostatectomy and had whole-mount processing of their prostate with careful tumor volume assessment. 34 What we found startled us; even in our military equal-access health care system, the black men had much higher tumor volumes overall and within each clinical stage. This within-stage tumor volume disparity was primarily responsible for the racial difference in PSA and PSA was a surrogate for bigger tumors in black men. Since this report, a large number of studies have con rmed that black American males, in general, have higher PSA values than white males. In particular, a number a recent studies from Vjayakumar and colleagues suggest that the racial disparity in PSA is primarily due to socioeconomics. 35 ± 36 In one investigation, blacks and whites who had similar insurance had similar PSA values; however, among Medicare bene ciaries blacks had higher PSAs. 35 Conversely, in another study they did nd that African-American men without prostate cancer had higher PSA and PSA density than whites or Hispanics after adjustment for age and prostate volume. 36 In a follow-up study of the same cohort, race was not an independent predictor of PSA in multivariable analysis and they concluded that PSA differences are due to sociological rather than biological causes. 37 In a multicenter registration study of the Radiation Therapy Oncology Group (RTOG), they also found that African-American men with nonmetastatic prostate cancer had higher PSA than whites and that household income, education and insurance alone or in combination did not completely account for this racial difference. 38 Finally, Vijayakumar et al rst described the `Will Rogers pheno-menon' in prostate cancer, that is, stage migration in the PSA-era for both blacks and whites. 39 Speci cally, they documented a 12.2% per year decline in median PSA values in newly diagnosed African-American men and concluded that widespread PSA screening was ameliorating the racial difference in PSA, proving the socioeconomic hypothesis. In contrast to this, our group recently updated our study of three-dimensional measured tumor volume and prediagnosis PSA values in a large cohort of military health care bene ciaries undergoing radical prostatectomy. 40 In 226 patients (155 whites and 46 blacks) treated between 1993 and 1997, blacks had signi cantly higher PSA values at diagnosis despite multivariable adjustment for tumor volume, benign gland volume, age, pathological stage and Gleason grade. Even in an equal access health care system with careful adjustment for comprehensively measured exact tumor burden, the African- American men had higher PSA, implying a biological basis. In a follow-up study, we also showed that this racial difference in PSA was not due to prostatitis or prostatic in ammation. 41 Conversely, Eastham et al found that it was more common for African-American men to exhibit in ammation on prostate biopsy. 42 Over time it could be possible that blacks exhibit more episodes of in ammation with more transient rises in PSA accounting for more prostate biopsies of African-American men with normal digital rectal examinations. 42 Still to be resolved as possible causes of higher observed PSA values in African-American men are greater amounts of high-grade prostatic intraepithelial neoplasia (PIN), 43 and higher PSA production or greater PSA `leakage' We do not believe that PIN is responsible in that PIN alone is not felt to bean source for PSA elevation. 44 Based on studies of serum testosterone and CAG repeats in the androgen receptor by race, 45,46 we believe that androgen stimulation may be responsible for greater PSA production in blacks. This hypothesis, however, must await further study. Proper use of PSA in African-American men Despite the widespread use of PSA, until the late 1990s no data existed to document the value of PSA testing and ` ne tune' the test for the early and accurate diagnosis of prostate cancer in this population. There was an urgent need to examine the proper PSA `normal' for black men and over the last few years a number of groups have examined the clinical utility of PSA in this population. Studies in particular by Smith and Catalona and colleagues from Saint Louis have shed much light in this area. 47 ± 49 In their rst study of white and 804 black men 50 y old, black men had a higher prevalence of elevated PSA ( 4.0 ng/ml, 13.1 vs 8.9%) and higher cancer detection rate (5.1 vs 3.2%). They also did not see a signi cantly higher cancer stage in the black screening volunteers who were diagnosed compared to whites but they did not draw signi cant volunteers from low SES zip code areas. In their second study, they found that a PSA > 4.0 ng/ml detected more cancer than digital rectal exam (DRE) for both blacks and whites. 48 Furthermore, the positive predictive value (PPV) for prostate cancer detection for both PSA and DRE was higher for blacks than whites (48 and 38% vs 38 and 22%). This study showed that the widely accepted PPV for PSA greater than 4.0 ng/ml of 25 ± 30% did not apply to African-American men who have a higher risk estimate of 36 ± 60%. In their most recent study, they lowered the threshold for PSA screening evaluation to 2.5 ng/ml and screened white men and 1004 black men 50 y old between 1995 and Blacks were younger (60 vs 63; P ˆ 0.005) and presented with higher PSA (3.3 vs 3.1 ng/ml; P ˆ 0.03) and black race remained an independent predictor of prostate cancer controlling for age, total PSA, PSA density, percentage free PSA, and number of prior screening visits. They concluded that lowering the screening threshold to 2.5 ng/ml was rea-

4 sonable, but felt that further study was needed to determine if this was the optimum cut-off point to screen African-American men. Our group has also studied the ability of PSA to detect prostate cancer in both Caucasian and African-American men and developed age-adjusted PSA reference ranges for maximal cancer detection in this high-risk group of men. 50 In this study, between January 1991 and May 1995 serum PSA concentration was determined for 3475 men without clinical evidence of prostate cancer (1802 Caucasian, 1673 African-American) and 1783 men with the disease (1372 Caucasian, 411 African-American). All PSA examinations were performed using Abbott IMx assay (normal 0 ± 4 ng/ml) in a central, single laboratory. PSA concentration was analyzed as a function of age and race to determine operating characteristics of PSA for the diagnosis of prostate cancer. Serum PSA concentration correlated directly with age for both black and white men (r ˆ 0.40, P ˆ for blacks and r ˆ 0.34, P ˆ for whites. African-American men had signi cantly higher serum PSA concentrations than Caucasian men (P ˆ Figure 1). When sensitivity was plotted against 1-speci city, the area under the receiver operator characteristic (ROC) curve was 0.91 for black men, and 0.94 for white men, indicating that the PSA test is an excellent early detection tool. For comparison, the Papanicolaou smear for cervical cancer, which is an accepted clinical screening test, has a ROC value of When we calculated age-speci c reference ranges by the identical methodology of Oesterling and colleagues used in their 1993 study of primarily Caucasian patients from Olmstead County, Minnesota, 51 we found very similar values for white men but higher values for black men. These ranges were 0 ± 2.4 ng/ml for black men aged 40 ± 49, 0 ± 6.5 ng/ml for men aged 50 ± 59, 0 ± 11.3 ng/ml for men aged 60 ± 69, and 0 ± 12.5 ng/ml for men aged 70 ± 79. We then tested these new ranges in our group of black men with prostate cancer to determine how these ranges would have performed if they had been used to detect their cancers. Unfortunately, these markedly higher ranges would have missed 41% of the cancers (only 59% sensitivity). The reason these traditionally derived ranges performed so poorly is that they are simply the 95th percentiles of values in the black controls. Because there is more variability of PSA results in blacks without evidence of cancer, there is more skewness, which pushes the 95th percentile farther to the right (higher). This higher range, however, is not clinically useful. We therefore developed age-adjusted reference ranges for black men with prostate cancer, selecting PSA upper limits of normal by decade to maximize cancer detection. In other words, we developed reference ranges by decade in the men with prostate cancer by using the 5th percentile of PSA values. Only the lowest 5% of pre-diagnosis PSA values in the black men with cancer are `normal' and the remainder (95%) are above the normal (95% sensitivity). We refer to these ranges as the Walter Reed/Center for Prostate Disease Research age-speci c reference ranges for maximal cancer detection (Table 1). They maximize sensitivity (cancer detection) without undue loss of speci city (false positive/unnecessary TRUS/biopsy). These values for maximal cancer detection for black and white men are compared to the traditional normal (0 ± 4 ng/ml) and the previously developed age-speci c reference ranges in Table 2. Table 1 Walter Reed/Center for Prostate Disease Research (CPDR) age-adjusted reference ranges of PSA for maximal prostate cancer detection in African-American patients African-American Decade PSA (ng/ml) Speci city 40 ± 49 0 ± % 50 ± 59 0 ± % 60 ± 69 0 ± % 70 ± 79 0 ± % From Morgan et al Figure 1 Distribution of PSA values for African-American (AA) and Caucasian (C) by age in decades for 3475 men without clinical evidence of prostate cancer (1802 C and 1673 AA). The line represents the 5th through 95th percentiles with the heavy horizontal bar being the 95th percentile. The boxes represent the 25th to 75th percentile and the mean is a line within the box. Note that AA men have more variability in PSA, which markedly increases the 95th percentiles compared to the C men. (Modi ed from Reference 50).

5 252 Table 2 Comparison of Walter Reed/Center for Prostate Disease Research PSA reference ranges for maximal prostate cancer detection vs traditional age-adjusted and original normal PSA value Walter Reed/Center for Prostate Disease Research (WR/CPDR) age-adjusted PSA reference ranges for maximum cancer detection: 50 Traditional PSA Traditional age-adjusted reference ranges based `normal' for on Caucasian men, Age African-American Caucasian all men zmayo Clinic ± 49 0 ± ± ± ± ± 59 0 ± ± ± ± ± 69 0 ± ± ± ± ± 79 0 ± ± ± ± 6.5 Our age- and race-adjusted PSA reference ranges have been met with some criticism. Littrup has been concerned that the average reader will not realize that our reference ranges are based on men with prostate cancer and are designed for sensitivity of cancer detection, not speci city. 52 He feels that we should refer to the ranges as `agesensitive reference ranges'. Furthermore, he feels that our values are perhaps too complex and would favor only two PSA `normals' > 2.0 ng/ml for `high-risk men' and > 4.0 ng/ml for the `general' population. 53 He is also concerned that any of these presently derived decision points for PSA only address `cancer' or no `cancer', not de ning a lower decision level for PSA in this high risk group to reduce their disproportionate mortality. Most recently, Powell and colleagues have also been critical of the Walter Reed/CPDR PSA reference ranges for African-American men. 54 They studied 651 consecutive radical prostatectomy patients who had a pretreatment PSA value and were operated between 1991 and They found that disease stage and grade was similar or worse for blacks compared to whites at all PSA ranges. Their basic concern was that we should not have raised the threshold PSA level to prompt a prostate biopsy above that for Caucasian men at any age because black men are at higher risk for cancer and have historically presented when their disease was more advanced. They feel that using a lower PSA threshold in black men may lessen the disparity that currently exists. On one hand, I agree with them. In fact, I have proposed lower PSA reference ranges to increase the odds of curable cancer in black men. 55 The problem is that one would need to biopsy all black men with a PSA greater than approximately 1.0 ng/ml to provide a 95th probability of curable cancer. 55 This illustrates the dilemma of various PSA thresholds to prompt prostate biopsy Ð the lower the cut-off point the higher the sensitivity (fewer missed cancers at potentially more curable stages) and the higher the cut-off point the better the speci city (fewer `unneces-sary' biopsies), but more missed cancers at potentially higher stages. Despite Powell et al's 54 concern about our proposed age-adjusted PSA values, 50 I do not feel that they have proven their point. In my opinion, the important question is: is there any difference in stage, grade or outcomes in the range where the PSA cut-off points are disputed? Speci cally, for men in their 50s is there a racial disparity between PSA of 3.5 and 4.0 ng/ml and between 3.5 and 4.5 ng/ml for men in their 60s. The authors found that African-American men had lower organ-con ned rates than Caucasian men when the pretreatment PSA was between 4.1 and 7.9, but this is not relevant to the range that they are disputing ie between 3.5 and 4.5 Furthermore, they showed signi cant racial disparity in PSArecurrence in men with PSA 4.0 and 10.0 but this is not relevant to the reference range debated between 3.5 and 5.5 ng/ml. Finally, their study did not address the ranges in question: 3.5 ± 4.0, 3.5 ± 4.5 and 3.5 ± 5.5 ng/ml. Their comparisons with all African-American men and Caucasian men above various PSA values are not valid. By including all subjects with PSAs above a certain cut-off point biases the analysis of survival. Speci cally, if there are more African-American men with high PSAs, who obviously will have a lower DFS, it will bias the statistics. Unfortunately, they did not have suf cient numbers of cases to adequately address whether any meaningful outcome differences by ethnicity were present in the narrow PSA cut-off point ranges that are debated. As previously noted, our group is attempting to develop PSA reference ranges for curable prostate cancer in African-American men. 56 Using the criteria of curability after radical prostatectomy de ned by Carter et al, from Johns Hopkins, 57 only 45% of contemporaryera black men who underwent a radical prostatectomy at our hospital were `curable'. 56 This compares to 74% for the predominantly white patients reported by Carter et al. 57 Furthermore, by pre-treatment PSA, Table 3 shows the striking racial differences in curability. Based on this, we are currently conducting a multicenter study to de ne age-adjusted PSA reference ranges for curable prostate cancer using age, pre-treatment PSA value and only those men who meet the curable criteria. Clinicians screening black (and white) men for prostate cancer currently have a wide choice of PSA thresholds to consider including the traditional 4.0 ng/ml for everyone, the Mayo Clinic age-adjusted ranges, 51 the Walter Reed/ Center for Prostate Disease Research ranges, 50 a 2.5 ng/ ml cut-off point along with percentage free PSA advocated by Catalona et al, 58 or the 2.0 ng/ml threshold advocated by Littrup from Wayne State for all African- American men. 52 Perhaps more important than `splitting hairs' about these subtle differences in PSA level, the key factor may be initiating screening at a younger age (40 ± 45 y) and following through with periodic sequential testing during a window of opportunity (ie 40 ± 60 or 45 ± 65 y, as an example).

6 Table 3 Comparison of curable prostate cancer by pre-treatment PSA value in black vs predominantly white radical prostatectomy patients from two institutions 253 Johns Hopkins a Walter Reed Number Racial group Predominantly white All black Era 1989 ± Ð 1996 Stage selection Clinically organ con ned nonpalpable Clinically organ con ned nonpalpable Curability b by pre-treatment PSA Value 34/36 (94%) 10/12 (83%) 4.0 ng/ml 32/36 (89%) 0/2 (0%) > 4.0 ± /317 (70%)* 43/104 (41%)* > 5.0 Overall 287/389 (74%) 53/118 (45%) a Carter et al. 57 b Curability de ned as organ-con ned with any grade or capsular penetration only (no margin, seminal vesicle or node positivity) with Gleason sum of < 7. *P-value. Despite the continuing controversy about the `exact' proper PSA by age and race, the most important concept, in my opinion, is recognizing that a PSA screening cut-off point of 4.0 ng/ml is probably too high for younger men, such as African-American men between 40 and 49 y of age. Bullock et al screened 214 black men between 40 and 49 y of age and found a prevalence of prostate cancer of 0.9% (2 out of 214) when a PSA of 4.0 ng/ml was used. 59 Interestingly, this prevalence increased to 5.6% (2 out of 36) when the black men also had a family history of prostate cancer. Conversely, Catalona et al from the same university, found that the cancer detection rate was 38% in a small group of 16 black men who had biopsy for PSA values between 2.6 and 4.0 ng/ml. 60 In a follow-up larger series, this same group found an even higher prevalence of 42% for African-American men with PSA between 2.6 and 4.0 ng/ml. 5 Our Department of Defense-funded Center for Prostate Disease Research (CPDR) has recently conducted three studies that illustrate that a PSA of 4.0 ng/ml is signi cantly higher than normal for young men. In 750 black and 750 white military members between the ages of 15 and 45 who had serum banked in the Department of Defense Serum Repository (DoDSR), the mean PSA values were 0.52 and 0.47 ng/ml, respectively. 61 The 95th percentile ranged from 1.16 to 1.38 ng/ml for the blacks strati ed by decade of age (1.38 ng/ml in the 40 ± 49 group); for the whites the corresponding values were 0.71 ± 1.13 ng/ml. Based on this new data, even using a PSA value of 2.0 ng/ml as a cut-off point for both black and white men is considerably higher than these 95th percentile values. Our second study is a prospective screening study of healthy of cers enrolled in the US Army War College at Carlisle Barracks, PA and is a collaboration between CPDR and the Army Physical Fitness Research Institute. 62 Between 1997 and 1999, 602 otherwise healthy military of cers (86.2% Caucasian) age 40 ± 49 y had a PSA test as part of a comprehensive health assessment at enrollment. In this primarily white cohort, only 10 of 602 (1.7%) had PSA 2.5 ng/ml and only 2 (0.5%) had PSA 4.0 ng/ml. All 10 men who had PSA 2.5 had sextant prostate biopsy and only one patient (0.17%) has been diagnosed with prostate cancer to date. The mean PSA value for the 601 men without cancer was 0.73 ng/ml and the 95th percentile (screening cut-off point) was 1.6 ng/ml. We are initiating a similar study in a non-commissioned of cer military school to increase data on African American men between 40 and 49. Our third study is a retrospective study of 1105 Students aged 30 ± 59 y, from the National Defense University, Fort McNair, Washington, DC between 1994 and For men in the 30 ± 39, 40 ± 49, and 50 ± 59 age groups, the 95th percentile of PSA was 1.6, 2.3 and 2.7 ng/ ml, respectively. If 18 men (1.6%) who had PSA 4.0 were excluded, the corresponding PSA values are 1.6, 2.1 and 2.4, respectively. Unfortunately, this study did not have ethnic information, therefore, we could not stratify by race. These three studies clearly illustrate that PSA values are very low for the vast majority of younger men, both black and white. Using a lower screening threshold below 4.0 ng/ml in men in their 40s (and perhaps 50s) would appear to be justi ed. Conclusions African-American men are at higher risk for being diagnosed with prostate cancer, although the cause is unknown. Almost all ethnic comparisons to date have found that blacks present with more advanced prostate cancer and have higher mortality. Recent studies in the PSA era suggest that this disparity may be partially or completely eliminated by education, equal-access and similar screening compliance. In general, PSA values are higher in African-American men. While most of this difference is due to greater tumor burden in blacks, our research shows slightly higher PSA values in African- American men even with age, grade and tumor volume taken into account. This difference may have implications for PSA screening reference ranges although the `exact' proper screening guidelines remain hotly debated. Based on current knowledge, I believe that using a lower PSA screening threshold of 2.0 ng/ml and educating

7 254 and testing high-risk African-American men annually starting at age 40 is justi ed. 64 References 1 Boring CC, Squires TS, Health CW. Cancer statistics for African- Americans. CA Cancer J Clin 1993; 43: 7 ± Kosary CL et al. SEER cancer statistics review, 1973 ± 1992: table and graphs. National Cancer Institute. National Institutes of Health publication no , Morton RA. Racial differences in adenocarcinoma of the prostate in North American men. Urology, 1994; 44: 637 ± Aronson WJ, Freeland ST. Editorial: can we lower the mortality of black men with prostate cancer? JUrol2000; 163: 150 ± Cooper R, David R. The biologic concept of race and the application to public health and epidemiology. J Health Polit Policy Law 1986; 11: 97 ± Osborne NG, Feit MD. The use of race in medical research. JAMA 1992; 267: 275 ± Witzig R. The medicalization of race: scienti c legitimization of a awed social construct. Ann Intern Med 1996; 125: 675 ± Brawley OW, Freeman HP. Race and outcomes: is this the end of the beginning for minority health research? J Natl Cancer Inst 1999; 91: 1908 ± Robinson SB, Ashley M, Haynes MA. Attitudes of African Americans regarding screening for prostate cancer. J Natl Med Assoc 1996; 88: 241 ± Robinson SB, Ashley M, Haynes MA. Attitudes of African Americans regarding prostate cancer clinical trials. J Commun Health 1996; 21: 77 ± Abbott RR, Taylor DK, Barber K. A comparison of prostate knowledge of African American and Caucasian men: changes from prescreening baseline to post intervention. Cancer J Sci Am 1998; 4: 175 ± Bennett CL et al. Relation between literacy, race, and stage of presentation among low-income patients with prostate cancer. J Clin Oncol 1998; 16: 3101 ± Guidry JJ, Fagan P, Walker V. Cultural sensitivity and readability of breast and prostate printed cancer education materials targeting African Americans. J Nat Med Assoc 1998; 90: 165 ± Myers RE et al. Adherence by African American men to prostate cancer education and early detection. Cancer 1999; 86: 88 ± Weinrich SP et al. Increasing prostate cancer screening in African American men with peer-educator and client-navigator interventions. JCancerEduc1998; 13: 213 ± Powell IJ et al. Outcomes of African American men screened for prostate cancer. The Detroit Education and Early Detection study. JUrol1998; 158: 146 ± Moul JW. Editorial comment to Outcomes of African American men screened for prostate Cancer: the Detroit Education and Early Detection Study. JUrol1997; 158: Mouton CP. Special health considerations in African-American elders. Am Family Physician 1997; 55: 1243 ± McCoy-Sibley RI, Sibley AF. Correlation of digital rectal examination, prostate speci c antigen, and transrectal ultrasound in prostate carcinoma in African Americans. J Nat Med Assoc 1997; 89: 318 ± Woolf SH. Screening for prostate cancer with prostate-speci c antigen. An examination of the evidence. New Engl J Med 1995; ± Barry MJ, Robert RG. Indications for PSA testing. JAMA 1997; 277: American College of Physicians. Screening for Prostate Cancer. Ann Internal Med 1997; 126: 480 ± Von Eschenbach A et al. American Cancer Society guidelines for the early detection of prostate cancer. Cancer 1997; 80: 1805 ± Hankey BF et al. Cancer surveillance series: Interpreting trends in prostate cancer Ð Part I: evidence of the effects of screening in recent prostate cancer incidence, mortality and survival rates. J Nat Cancer Inst 1999; 91: 1017 ± Etzioni R et al. Cancer surveillance series: Interpreting trends in prostate cancer Ð Part III: quantifying the link between population prostate-speci c antigen testing and recent declines in prostate cancer mortality. J Natl Cancer Inst 1999; 38: 1033 ± Labrie F et al. Screening decreases prostate cancer death: First analysis of the 1998 Quebec prospective randomized controlled trial. Prostate 1999; 38: 83 ± Roach J et al. The prognostic signi cance of race and survival from prostate cancer based on patients irradiated on Radiation Therapy Oncology Group Protocols (1976 ± 1985). Int J Radiat Oncol Bid Phys 1992; 24: 441 ± Optenberg SA et al. Race, treatment, and long-term survival from prostate cancer in an equal-access medical care delivery system. JAMA 1995; 274: Eastham JA, Kattani MW. Disease recurrence in black and white men undergoing radical prostatectomy for clinical stage T1 ± T2 prostate cancer. J Urol 2000; 163: 143 ± Powell IJ et al. Should African American men be tested for prostate carcinoma at an earlier age than white men? Cancer 1999; 85: 472 ± Vijayakumar S et al. Racial differences in prostate-speci c antigen levels in patients with local-regional prostate cancer. Cancer Epidemiol Biomarkers Prev 1993; 1: 541 ± Staggers FM et al. A pro le of African-American participants in Prostate Cancer Awareness Week (PCAW) J Urol 1994; 151: 291(A#255). 33 Bullock A et al. Racial differences in prostate cancer detection and staging. JUrol1994; 151: 127 ± Moul JW et al. Prostate-speci c antigen values at the time of prostate cancer diagnosis in African-American men. JAMA 1995; 274: 1277 ± Vijayakumar S et al. Prostate speci c antigen levels in African Americans correlate with insurance status as an indicator of socioeconomic status. Cancer J Sci Am 1996; 2: 225 ± Abdalla I et al. Comparison of serum prostate-speci c antigen levels and PSA density in African American, white, and Hispanic men without prostate cancer. Urology 1998; 51: 300 ± Abdalla I, Ray P, Vaida F, Vijayakumar S. Racial differences in prostate-speci c antigen levels and prostate speci c antigen densities in patients with prostate cancer. Am J Clin Oncol 1999; 22: 537 ± Vijayakumar S et al. Prostate speci c antigen levels are higher in African Americans than in white patients in a multicenter registration study: results of RTOG Int J Radiat Oncol Biol Phys 1998; 40: 17 ± Vijayakumar S, Vaida F, Weichselbaum RR, Hellman S. Race and the Will Rogers phenomenon in prostate cancer. Cancer J Sci Am 1998; 4: 27 ± Moul JW et al. Racial differences in tumor volume and prostate speci c antigen among radical prostatectomy patients. J Urol 1999; 162: 394 ± Zhang W et al. In ammatory in ltrate (prostatitis) in wholemounted radical prostatectomy specimens from black and white patients is not an etiology for racial difference prostatespeci c antigen. JUrol2000; 163: 131 ± Eastman JA et al. Clinical characteristics and biopsy specimen features in African American and white men without prostate cancer. J Natl Cancer Inst 1998; 90: 756 ± Sakr WA et al. Epidemiology of high grade intraepithelial neoplasia. Path Res Prac 1995; 191: 838 ± Sakr WA et al. Inter-national consultation on prostatic intraepithelial neoplasia and pathologic staging of prostatic carcinoma: staging and reporting of prostate cancer sampling of the radical prostatectomy specimen. Cancer 1996; 78: Ross RK et al. 5- alpha-reductase activity and risk of prostate cancer among Japanese and US white and black males. Cancer 1992; 339: Irvine RA, Ross RK, Coetzee GA. The CAG and GGC microsatellites of the androgen receptor gene are in linkage disequilibrium in men with prostate cancer. Cancer Res 1995; 55: Smith DS, Bullock AD, Catalona WJ, Herschan JD. Racial differences in a prostate cancer screening study. J Urol 1996; 156: 366 ± 369.

8 48 Smith DS, Bullock AD, Catalona WJ. Racial differences in operating characteristics of prostate cancer screening tests. J Urol 1997; 158: 1861 ± Smith DS et al. Use of lower prostate-speci c antigen cut-offs for prostate cancer screening in black and white men. JUrol1998; 160: 1734 ± Morgan TO et al. Age-speci c reference ranges for prostatespeci c antigen in black men. New Engl Med 1996; 335: 304 ± Oesterling JE et al. Serum prostate-speci c antigen in a community-based population of healthy men. JAMA 1993; 270: Littrup PJ, Editorial: Prostate cancer in African-American men. Prostate 1997; 31: 139 ± Littrup PJ, Sparschu RA. Transrectal ultrasound and prostate cancer risks: the `tailored' prostate biopsy. Cancer 1995; 75(Suppl): 1805 ± Powell IJ et al. Should the age-speci c prostate speci c antigen cutoff for prostate biopsy be higher for black than for white men older than 50 y? JUrol2000; 163: 146 ± Moul JW. Curability of prostate cancer in African Americans: implications for neoadjuvant/adjuvant hormonal therapy. Mol Urol 1998; 2: Moul JW. PSA thresholds for prostate cancer detection. JAMA 1997; Carter HB et al. Recommended prostate-speci c antigen testing intervals for the detection of curable prostate cancer. JAMA 1997; 277: 1456 ± Catalona WJ. Screening for prostate cancer (Letter). New Eng J Med 1996; 334: 666 ± Bullock AD, Harmon T, Smith DS, Basler JW. Prostate cancer screening in younger men. JUrol1997; 157(Suppl 66, Abstract 252). 60 Catalona WJ, Smith DS, Ornstein DK. Prostate cancer detection in men with serum PSA concentrations of 2.6 to 2.4 ng/ml and benign prostate examination. JAMA 1997; 277: 1452 ± Preston DM, Levin LI, Jacobson. et al. Prostate speci c antigen levels in young Caucasian and African American men age 20 ± 45. Urology (in press). 62 Moul JW, Connelly RR, Barko WF, Vaitkus M. Should healthy Caucasian men between the age of 40 ± 49 be screened for prostate cancer: a Department of Defense (DoD) Center for Prostate Disease Research (CPDR) and Army Physical Fitness Research Institute (APFRI) prospective study at the U.S. Army War College (USAWC). J Urol 2000; 163(Suppl 4): 90 (abstract 393). 63 Hartzell J, Kao C, Moul JW. Age-adjusted prostate-speci c antigen (PSA) in men between 30 ± 59 y of age: National Defense University study. The Prostate J (in press). 64 Moul JW. A PSA cutoff of 4 is too high for many men. Contemp Urol 1999; 11: 15 ±